From owner-freebsd-hackers Mon Jul 30 14: 9:46 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from elvis.mu.org (elvis.mu.org [216.33.66.196]) by hub.freebsd.org (Postfix) with ESMTP id 0E3F937B401 for ; Mon, 30 Jul 2001 14:09:44 -0700 (PDT) (envelope-from bright@elvis.mu.org) Received: by elvis.mu.org (Postfix, from userid 1192) id E155A81D05; Mon, 30 Jul 2001 16:09:43 -0500 (CDT) Date: Mon, 30 Jul 2001 16:09:43 -0500 From: Alfred Perlstein To: Steven Ames Cc: Jaye Mathisen , hackers@FreeBSD.ORG Subject: Re: Softupdate gripe... Message-ID: <20010730160943.S26571@elvis.mu.org> References: <20010730132122.C548@apocalypse.cdsnet.net> <20010730160043.Q26571@elvis.mu.org> <023001c1193a$d1de9f30$28d90c42@eservoffice.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <023001c1193a$d1de9f30$28d90c42@eservoffice.com>; from steve@virtual-voodoo.com on Mon, Jul 30, 2001 at 04:01:31PM -0500 Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG * Steven Ames [010730 16:04] wrote: > > ----- Original Message ----- > From: "Alfred Perlstein" > > * Jaye Mathisen [010730 15:21] wrote: > > > > > > 2 500+GB FS's, both filled completely. > > > > > > Filesystem 1K-blocks Used Avail Capacity > Mounted on > > > /dev/twed3d 524698116 524697730 386 100% > /d/d0 > > > /dev/twed4d 524698116 524502226 195890 100% > /d/d1 > > > newsfeed-inn# rm -f /d/d?/* > > > newsfeed-inn# ls -lR /d > > > total 4 > > > I hope you're not expecting that 'rm' command to somehow realize that > > it's operating on two seperate FS's. If you want to see deletions > > happen at the same time you want to do this: > > The lack of an ampersand on this command: 'newsfeed-inn# rm -f /d/d?/*' > rather had me believe he waited for it to complete before starting to do > his 'df' and was suprised at how long it was taking softupdates to release > diskspace and at the order in which it was doing it. Oh, then he just needs to issue a 'sync'. Enough pressure on the memory/disk subsystem (or false pressure from sync(1)) should cause it round robin, thing is that since the intial removal was serialized normal LRU behavior is to process whatever was removed first. -- -Alfred Perlstein [alfred@freebsd.org] Ok, who wrote this damn function called '??'? And why do my programs keep crashing in it? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message